RESEARCH ARTICLE

IDOL N342S Variant, AtherosclerosisProgression and Cardiovascular Disorders inthe Italian General PopulationAshish Dhyani1☯, Gianpaolo Tibolla1,2☯, Andrea Baragetti3, Katia Garlaschelli3,Fabio Pellegatta3, Liliana Grigore2,3, Giuseppe Danilo Norata1,3,4*, AlbericoLuigi Catapano1,2*

1 Department of Pharmacological and Biomolecular Sciences, University of Milan, Milan, Italy, 2 I.R.C.C.SMultiMedica, Milan, Italy, 3 SISA Center for the Study of Atherosclerosis, Bassini Hospital, Cinisello B., Italy,4 The Blizard Institute, Centre for Diabetes, Barts and The London School of Medicine & Dentistry, QueenMary University, London, United Kingdom

☯ These authors contributed equally to this work.* danilo.norata@unimi.it (GDN); alberico.catapano@unimi.it (ALC)

AbstractInducible degrader of the low density lipoprotein receptor (IDOL), is an E3 ubiquitin ligase

that negatively modulates low density lipoprotein receptor (LDL-R) expression. Genome-

wide association studies (GWAS) indicated that genetic variants in IDOL gene contributes to

variation in LDL-C plasma levels and the detailed analysis of a specific locus resulted in the

identification of the functional common single nucleotide polymorphism (SNP) rs9370867 (c.

G1025A, p.N342S) associates with increased LDL-R degradation and increased LDL-C lev-

els. These findings, however, were not confirmed in two other independent cohorts and no

data about the impact of this variant on atherosclerosis progression and cardiovascular risk

are available. Aim of this study was to investigate the association between a functional variant

in IDOL and atherosclerosis progression in an Italian general population. 1384 subjects en-

rolled in the PLIC study (Progression of Lesions in the Intima of Carotid) were genotyped by

Q-PCR allelic discrimination and the association with anthropometric parameters, plasma lip-

ids and the carotid intima media thickness (cIMT) and the impact on cardiovascular disease

(CVD) incidence were investigated. The N342S variant was not associated with changes of

the plasma lipid profile among GG, AG or AA carriers, including total cholesterol (249±21,

249±19 and 248±21 mg/dl respectively), LDL-C (158±25, 161±22 and 160±23 mg/dL), cIMT

(0.74±0.14, 0.75±0.17 and 0.77±0.15 mm) and CVD incidence. In agreement, the expression

of LDLR and the uptake of LDL was similar in macrophages derived fromGG and AA carriers.

Taken together our findings indicate that the N342S variant does not impact plasma lipid pro-

file and is not associated with atherosclerosis progression and CVD in the general population,

suggesting that other variants in the IDOL genemight be functionally linked with cholesterol

metabolism.

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 1 / 12

OPEN ACCESS

Citation: Dhyani A, Tibolla G, Baragetti A,Garlaschelli K, Pellegatta F, Grigore L, et al. (2015)IDOL N342S Variant, Atherosclerosis Progressionand Cardiovascular Disorders in the Italian GeneralPopulation. PLoS ONE 10(4): e0122414.doi:10.1371/journal.pone.0122414

Academic Editor: Ramona Natacha PENA iSUBIRÀ, University of Lleida, SPAIN

Received: November 7, 2014

Accepted: February 20, 2015

Published: April 30, 2015

Copyright: © 2015 Dhyani et al. This is an openaccess article distributed under the terms of theCreative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the original author and source arecredited.

Data Availability Statement: Individual-level dataare not publicly available due to patient privacyconcerns. Raw data can be requested from:centroatero.cat@gmail.com.

Funding: The authors have no support or funding toreport.

Competing Interests: Alberico Catapano andGiuseppe Norata are PLOS ONE Editorial Boardmembers. This does not alter the authors' adherenceto PLOS ONE Editorial policies and criteria.

IntroductionElevated levels of circulating low density lipoprotein cholesterol (LDL-C) represent a key factorfor cardiovascular disease (CVD) risk [1]. Plasma LDL-C levels are mainly regulated by theproduction and the clearance of apolipoprotein B (apoB) containing lipoproteins by the liver.A central role in the hepatic LDL-C metabolism is played by the low density lipoprotein recep-tor (LDLR) that mediates the uptake of LDL in the hepatocytes, thus promoting their clearance[2]. LDLR gene mutations account for the majority of the cases of autosomal dominant hyper-cholesterolemia (ADH 1), a genetic disease characterized by elevated LDL-C levels and prema-ture CVD death [3]. The LDLR activity is controlled at the transcriptional level by the nucleartranslocation of the sterol regulatory element binding protein 2 (SREBP2) [4]. LDL-R expres-sion is also controlled by the pro-protein convertase subtilisin-like kexin type 9 (PCSK9) whichbinds the LDLR at the plasma membrane of the hepatocytes and induces its degradation in thelysosomes [5]. Circulating PCSK9 is mainly produced by the liver and in analogy with theLDLR is controlled by SREBP2 activity, thus making PCSK9 an interesting target for the devel-opment of lipid lowering drugs [6]. In addition to PCSK9, the inducible degrader of the LDLR(IDOL, also known as MYLIP) also controls the LDLR abundance by mediating the ubiquitina-tion of the intracellular tail of the receptor and its lysosomal degradation [7]. In contrast toPCSK9, IDOL expression is ubiquitous and regulated by the oxidized sterols sensitive nuclearreceptor liver X receptor (LXR) [8]. Hepatic overexpression of the IDOL gene in mice results inhypercholesterolemia and atherosclerosis development [7,9]. In humans IDOL has been sug-gested as a candidate gene involved in the modulation of lipoproteins metabolism by genome-wide association studies (GWAS)[10–12]. Further investigation aimed at identifying the IDOLgenetic variants responsible for the association generated controversial results. In a Mexicandyslipidemic population fine mapping of the IDOL gene identified the common rs9370867SNP as the susceptibility variant associated with total cholesterol levels [13]. The rs9370867SNP encodes the amino-acid substitution N342S, located in the FERM domain of the protein,a critical region involved in the regulation of protein-protein interaction. The presence of a Ser-ine residue (encoded by the G allele) reduces the ability of the IDOL protein to ubiquitinate theLDLR, thus increasing plasma membrane LDLR expression and LDL clearance. This observa-tion was not replicated in two Brazilian cohorts, characterized by mixed ethnicity, where thesame variant was not associated with LDL-C levels both the general population and in patientswith stable angina [14]. Finally, analysis of IDOL gene variants in the Dutch populationshowed similar allelic frequencies of the rs9370867 SNP in two cohorts with extremely highand low LDL-C levels [15], further questioning the functional relevance of this variant. In thisstudy we aimed at clarifying the contribution of the rs9370867 SNP in the IDOL gene to plasmacholesterol levels and the association with a preclinical marker of atherosclerosis such as thecommon carotid artery intima media thickness (cIMT). For this purpose we genotyped a co-hort representative of the Italian general population enrolled in the PLIC study (Progression ofLesions in the Intima of Carotid- Caucasian ethnicity)[16–18] and we investigated the associa-tion of the rs9370867 SNP with the lipid profile, cIMT, and the incidence of cardiovascular dis-ease (CVD) after a 10 years follow-up. We found no association of plasma lipid profile withcIMT and CVD incidence. Moreover we observed similar LDL uptake capacity in macrophagesgenerated from peripheral blood mononuclear cells from IDOL rs9370867 GG or AA carriers.Taken together our results indicate that other variants, if any, are likely to be responsible forthe relationship between plasma LDL-C levels and genetic variation in the IDOL gene evi-denced by recent GWAS in Caucasian populations.

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 2 / 12

Materials and Methods

Population sampleA cohort of 2141 subjects attending the Atherosclerosis Center in Bassini Hospital, Departmentof Pharmacological and Biomolecular Sciences (University of Milan, Italy), were recruited forthe PLIC study. The PLIC study was approved by the Ethics Commitee of the University ofStudy of Milan (approved on 06-02-2001 SEFAP protocollo n°0003/2001), all participants signeda written informed consent. The investigation was performed in accordance with the principlesof the Declaration of Helsinki. This project is a study designed to investigate the presence andprogression of atherosclerotic lesions and intima media thickness (IMT) in the carotid artery ofa large local cohort in relation to the major cardiovascular disease risk factors. Exclusion criteriawere use of hypolipidemic drugs, presence of liver or kidney disease, thyroid dysfunction. 1384subjects gave their signed consent to use their DNA for genetic studies addressing the cardiovas-cular system. Biochemical and clinical variables were evaluated as previously described [19,20].High resolution B-mode ultrasonography of carotid Intima-Media Thickness (c-IMT) with a lin-ear ultrasound probe (4.0–13.0 MHz frequency, 14X48 mm footprint, 38 mm field of view) wasperformed (Vivid S5 GE Healthcare1, Wauwatosa, WI, USA). The determination were per-formed by a single sonographer, blinded to the subject’s identity (intra‐class correlation = 0.812,n = 30). All the measurements were done off-line using the software provided by the instrument[21]. The protocol involved the common carotid artery (CCA) (30 mm proximal to the carotidbulb), the carotid bulb and internal carotid artery (ICA) at both sides. The intima-media thick-ness (IMT) was assessed at the far wall as the distance between the interface of the lumen and in-tima, and the interface between the media and adventitia in a standardized number of points[21,22]. The maximal IMT was recorded and averaged for the left and right sides of the CCA(30 mm proximal to the carotid bulb), the carotid bulb, the ICA. Presence of extra-cardiac ath-erosclerotic vascular involvement was determined with presence of focal plaques (> 1.3 mm inlongitudinal resolution, lateral or medial angle) and/or diffusive mean IMT> 1.3 mm (in longi-tudinal resolution, lateral or medial angle)[23].

The protocol involved the determination of the Left Ventricular Mass (LVM) assessment.The measurements were performed by two dimensional guided M-mode echocardiographyfrom the parasternal window using an M-mode color-Doppler (Vivid S5 GE Healthcare1,Wauwatosa, WI, USA) (1.4–4.0 MHz frequency, 19.3X27.6 mm footprint wide-band phasedarray transducer). Left Ventricular Mass (LVM) was calculated with Devereux’s formula, ac-cording to guidelines. The definition of cardiovascular events (CVEs) included coronary heartdisease (CHD), such as acute myocardial infarction (AMI), acute coronary syndromes, acuteand chronic heart failure (New York Heart Association class II and III); peripheral artery dis-ease (PAD) and cerebrovascular events, such as stroke, transient cerebral ischemic attack in theprevious 6 months, and other vascular complications (including diabetic foot ulcers). In addi-tion having undergone major surgery was included in this definition (i.e.: carotid thrombo-arterectomy, percutaneous coronary angioplasty, arterial angioplasty or arterial by-pass of thelower limbs, coronary by-pass and amputations). Cardiovascular risk was defined in accor-dance to the Progetto Cuore individual risk score, for the Italian population as described [23].

GenotypingGenomic DNA was extracted from buffy coat samples using the Flexigene DNA kit (Qiagen,Milan, Italy) according to the manufacturer’s instructions. Genotyping for the rs9370867IDOL SNP (c.G1025A, p.N342S) was performed on 5 μL (10–200 ng of DNA), using a TaqMan

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 3 / 12

allelic discrimination test. Taqman SNP Genotyping assay Code N. C_2461770_10 (ThermoFisher Scientific, Applied Biosystems, Waltham, MA).

Peripheral blood mononuclear cells and macrophages cell culturePeripheral blood mononuclear cells (PBMCs) were collected from 10 patients previously en-rolled in the PLIC study (n = 5 carriers of the AA genotype and n = 5 carriers of the GG geno-type, the two groups were matched for sex and age) as reported [24]. Briefly blood was diluted1: 3 in Phosphate Buffered Saline (PBS) (15 mL) then layered onto 4 mL of Ficoll Hipaque(Amersham) and centrifuged at 1500 rpm for 35 min. PBMCs were removed from the interfaceand washed twice in PBS before being re-suspended in RPMI-1640 medium supplementedwith penicillin (50 U/ml), streptomycin (50 μg/ml), L-glutamine (2 mM) and 10% serum AB.PBMCs were counted and 2x106 cells were plated in 6 well plates and incubated for 2 hours ina humidified atmosphere (37°C, 5% CO2). Non adherent cells were removed with four rinses ofPBS and attached monocytes cultured for 7 days to obtain monocyte-derived macrophages.

Gene expression analysisTotal RNA was extracted from monocytes-derived macrophages and 1 μg of RNA underwentreverse transcription using the IScript cDNA Synthesis kit (BioRad, Milan, Italy) [24,25].Three μL of cDNA were amplified by real-time quantitative PCR with 2X MAXIMA SYBRGreen/Fluorescein qPCR mastermix (Carlo Erba Reagents, Cornaredo, Italy)[26]. The specific-ity of the Sybr green fluorescence was tested by plotting fluorescence as a function of tempera-ture to generate a melting curve of the amplicon. Each sample was analyzed in duplicate usingthe CFX Connect Real Time detection system (BioRad, Milan, Italy). The primers used are thefollowing: 18S: Fw 5’-CGCAGCTAGGAATAATGGAATAGG-3’, Rw5’- CATGGCCTCAGTTCCGAAA-3’; LDLR: Fw 5’- GTGTCACAGCGGCGAATG -3’, Rw 5’- CGCACTCTTTGATGGGTTCA -3’. IDOL: Fw 5’- GATAACAGAGACGCACGCATTC -3’, Rw 5’- CCCTTCAAGTCACGGCTATACTG -3’.

LDL preparation and macrophage LDL uptake assayLDL (density 1.019–1.063 g/ml) were obtained from freshly isolated human plasma fromhealthy volunteers by preparative ultracentrifugation in KBr gradient, dialyzed versus PBS con-taining 0.01% EDTA and sterilized by filtration. Protein content was determined by the colori-metric Lowry assay, using BSA as a standard. We then used a flow-cytometry assay to evaluateLDL uptake by using fluorescently labelled LDL, a widely and robust assay currently used forthe investigation of LDLR function [27,28]. 1mg of LDL was labelled overnight at 4°C in 1 mLPBS containing 0,5mg/mL of 3,30-dioctadecyloxacarbocyanine perchlorate (DiO) (stock solu-tion 5mg/mL in dimethylformamide) (Sigma Aldrich, Milan, Italy). DiO-LDL were then isolat-ed again by ultracentrifugation in KBr gradient and the excess of fluorescent probe and KBrsalt was removed by dialysis versus PBS containing 0.01% EDTA. LDL were sterilized by filtra-tion and the protein content determined as previously described. Monocytes-derived macro-phages were washed with PBS then incubated with 10μg/mL of DiO-LDL in RPMI withoutserum

After 2 hours cells were washed three times with PBS, gently scraped off and collected intubes for flow cytometry analysis. Cells were immediately analyzed using a FACSCalibur flowcytometer (BD Biosciences) and CellQuest software.

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 4 / 12

Statistical analysisData were analyzed using IBM-SPSS 21.0 for Windows (Chicago, IL, USA). Results are re-ported as mean ± SD, if not otherwise stated; Grubb’s test was performed to verify presence ofoutliers data. Group differences among genotypes were determined by using one way Analysisof Variances (ANOVA) test followed by LSD and Bonferroni post-hoc analyses; otherwiseKruskal-Wallis test was then performed to compare not normally distributed variables acrossthe three genotypes. Then the Multivariable Analysis of Variances (MANOVA) was performedto adjust the comparisons with other covariates, if significantly different among genotypes. Stu-dent’s t-test for continuous normally-distributed variables (when appropriate) and χ2 analysisfor categorical variables were performed for the analysis of the alleles. Group differences withP< 0.05 were deemed as statistically significant.

Results

General characteristics of subjects from the PLIC population accordingto the rs9370867 SNPNo deviation from the Hardy-Weinberg equilibrium was observed for the rs9370867 (c.G1025A) SNP. In the PLIC population analyzed (n = 1384) the allelic frequencies were 48.01%and 51.99% for the G and A allele respectively, no significant differences in the genotypes dis-tribution between men (n = 485) and women (n = 899) were observed (Table 1). No major dif-ferences were found in the anthropometric and biochemical parameters, such as totalcholesterol (TC), LDL-C, high density lipoprotein cholesterol (HDL-C), triglycerides (TG) andglycaemia among carriers of the different genotypes. (Table 2). Previous association of thers9370867 SNP with cholesterol metabolism were observed in a Mexican dyslipidemic popula-tion, characterized by high levels of TC and TG. For this reason we stratified the PLIC popula-tion according to TG and TC plasma levels and we repeated the analysis in two sub-groupsupper the 75th percentile of TC (mean value 271.2 ± 23.7 mg/dL) and TG (mean value191 ± 71,1mg/dL). The results were similar compared to the whole population and showed nosignificant impact of the rs9370867 (c.G1025A) SNP on the lipid profile, S1A and S1B Table

Impact of the rs9370867 SNP for IDOL on carotid atherosclerosis andthe prevalence of cardiovascular eventsAlthough no association between the rs9370867 SNP and LDL-C or other lipid variables wereobserved, this does not exclude the possibility that this functional SNP might affects otherpathways associated with cardiovascular disease. Therefore to further investigate the potentialrole for the rs9370867 SNP, its association with left ventricular mass and intima media thick-ness of carotid arteries (Fig 1A and Fig 1B) and the incidence of coronary heart disease,

Table 1. Genotypic [n(%)] and allelic frequencies (%) of the rs9370867 SNP in the PLIC population.

Genotype Allele

n GG GA AA G A

PLIC 1384 328(23.7%) 673(48.6%) 383(27.7%) 48.01% 51.99%

Men 485 127(26.2%) 230(47.4%) 128(26.4%) 49.9% 50.1%

Women 899 201(22.4%) 443(49.3%) 255(28.4%) 47% 53%

χ2 2.621 2.552

P-value 0.454 0.110

doi:10.1371/journal.pone.0122414.t001

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 5 / 12

peripheral artery disease and cardiovascular events (CHD, PAD and CVE) was investigated(Fig 2A to Fig 2C). The results show that the rs9370867 genotype is not associated with any ofthe parameters investigated. Also when carriers of the A allele both in homozygosis and in het-erozygosis were compared to carriers of the G allele, no differences in the odd ratio for the inci-dence of CHD, PAD or CVE were observed. These data limit the impact of rs9370867 SNP forIDOL on cardiovascular outcome.

LDLR expression and activity according to the rs9370867 SNPThe relevance of the rs9370867 IDOL SNP on LDL-R expression was further investigated inmacrophages generated from peripheral blood mononuclear cells from N342 or S342 IDOLcarriers. LDL-R expression and the capacity to internalize LDL particle, as a functional readoutof the LDL-R activity. LDL-R mRNA expression and IDOL mRNA expression were similar inN342 and S342 macrophages (Fig 3A) and more importantly the genotype was not associatedwith differences in LDLR function. N342 and S342 macrophages showed indeed the same lev-els of LDL uptake, as assessed by flow cytometry after incubation of the cells with fluorescentlylabelled LDL particle (Fig 3B and Fig 3C).

DiscussionGWAS study in European descent, showed that MYLIP (IDOL) genetic loci are associated withplasma lipid levels [10–12]. The rs9370867 variant on the IDOL gene was initially proposed asa gain of function variant associated with increased LDL-cholesterol levels [13]; however subse-quent papers questioned the association of this variant with increased plasma total cholesterollevels both in Brazilians [14] and in Caucasians [15].

For this reason, we investigated in a large cohort from the Italian general population the im-pact of the same variant of the IDOL gene on plasma lipid parameters and on marker of pe-ripheral atherosclerosis. In agreement with studies in Brazilian and Dutch cohorts [14,15], theIDOL SNP was not associated with any relevant difference on plasma lipid profiles. Of note,while in Europeans the rs9370867 SNP variant is in strong linkage disequilibrium with two up-stream genome-wide significant GWAS variants rs2294261 [11], and rs2480 [12], this is nottrue in the Mexican population [13]. This finding raises the possibility that other SNPs couldexplain the relevance of the MYLIP locus in the control of plasma lipoprotein levels in Europe-ans [11] as opposed to what observed in the Mexican population. However neither rs2294261nor rs2480 exhibited a potential regulatory effect using cis-eQTL analysis [13], suggesting that

Table 2. Anthropometric and biochemical characteristics according to the rs9370867 SNP in subjects from the PLIC population.

GG AG AA P-value GG vs AA P-value GG vs GA P-value AA vs GA

Age 65.21 ± 9.97 64.36 ± 1.084 64.90 ± 9.88 0.735 0.342 0.515

Body mass index (Kg/m2) 26.8 ± 4.6 26.7 ± 4.3 27.0 ± 4.2 0.756 0.739 0.456

Total cholesterol (mg/dL) 209.9 ± 35.9 206.1 ± 34.6 204.7 ± 34.3 0.089 0.172 0.570

LDL-cholesterol (mg/dL) 125.2 ± 32.8 123.3 ± 31.9 122.6 ± 31.6 0.356 0.449 0.775

HDL-cholesterol (mg/dL) 64.60 ± 18 63.08 ± 16.18 62.64 ± 14.8 0.168 0.249 0.701

Triglycerides (mg/dL) 100.5 ± 42.4 98.91 ± 48.0 97.27 ± 45.6 0.391 0.646 0.634

Apolipoprotein A-I (mg/dL) 158.2 ± 23.4 156.4 ± 21.0 155.3 ± 20.7 0.133 0.311 0.455

Apolipoprotein B (mg/dL) 108.7 ± 22.1 105.6 ± 22.2 107.0 ± 22.2 0.389 0.078 0.380

SBP (mmHg) 128 ± 17 129 ± 18 128 ± 18 0.568 0.352 0.749

DBP (mmHg) 82.03 ± 9.4 82.37 ± 8.7 82.32 ± 8.8 0.983 0.228 0.195

Glucose (mg/dL) 95.38 ±15.89 94.97 ± 17.93 92.94±13.72 0.057 0.762 0.091

doi:10.1371/journal.pone.0122414.t002

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 6 / 12

other studies are need to identify the actual casual variants for these GWAS signals. Further-more the frequency of allele “A” and “G” alleles for rs9370867 variants of IDOL in our subjectswas similar to what reported in larger genome databases but differs from that reported byWeissglas- Volkov et al in the Mexican dyslipidemic cohort [13]. While these aspects supportthe importance of adjustment for ethnicity, when this was performed in the Brazilian cohort,again no significant association of the N342S variant was observed [14].

In the attempt to investigate whether, despite the absence of a significant association to plas-ma lipid levels, the rs9370867 variant on the IDOL gene impacts cardiovascular outcome, weinvestigated its association with cardiac and vascular damage. Carriers of the different allelesfor the rs9370867 variant on the IDOL gene present a similar degree of cardiac and vasculardysfunction and more importantly do not differ for the prevalence of either coronary heart dis-ease, peripheral vascular disease or cardiovascular events. Our observations further support alimited impact of this variant in Europeans and strongly support the need of validating the as-sociation with cardiovascular outcome also in the Mexican cohort.

Fig 1. Heart left ventricular mass and carotid intimamedia thickness of the common carotid arteriesin the PLIC population according to the rs9370867 IDOL SNP. Panel A shows the mean ± SD for leftventricular mass for GG, AG or AA carriers. Panel B shows the mean ± SD of carotid intima media thicknessof the common carotid arteries (right and left) for GG, AG or AA carriers.

doi:10.1371/journal.pone.0122414.g001

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 7 / 12

Fig 2. Incidence of coronary heart disease (CHD), peripheral artery disease (PAD) and cardiovascularevents (CVE) in the PLIC population according to the rs9370867 SNP. Panels A, B and C shows theincidence (%) of CHD, PAD and CVE for GG, AG or AA carriers of the rs9370867 IDOL SNP. Panel D showsthe risk CHD, PAD and CVE for the carriers of A allele of the rs9370867 vs carriers of the G allele. OddsRatios (OR, the 95% Confidence Interval (C.I.), adjusted for age, gender, lipid profile, systolic blood pressure,glucose levels and therapies) are not statistically significant.

doi:10.1371/journal.pone.0122414.g002

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 8 / 12

Based on the lack of association between the rs9370867 variant with plasma lipid variables,metabolic determinants as well as markers of cardiac and vascular dysfunction, we decided toinvestigate whether the N342S substitution affects IDOL function. Previous data, generated bycotransfecting human embryonic kidney 293T cells (HEK293T) with plasmids for LDLR andeither N342 IDOL or S342 IDOL showed that the presence of N342 is associated with increasedLDL-R receptor degradation [13] pointing toward a functional effect for the rs9370867 variant.This finding however was not confirmed in a different work, which was performed under thesame experimental conditions by overexpressing the LDL-R and N342S IDOL variant inHEK293T cells [15]. To shed more light on this conflicting result and given that IDOL is widely

Fig 3. mRNA expression of LDL-R and IDOL and LDL uptake in macrophages from N342 and S342macrophages. Panel A shows mRNA expression for LDL-receptor and for IDOL in macrophages obtainedfrom N342 and S342 carriers (n = 5 for both genotypes) cultured in complete RPMI medium The results,normalized for the expression of an housekeeping gene (18S), data are shown as mean ± SEM. Next flowcytometry was used to study the LDL uptake in macrophages obtained from N342 or S342 carriers. Todetermine the LDL uptake, macrophages were incubated with fluorescently labelled LDL (DiO-LDL, 10 μg/mL) for 2 hours at 37°C. Panel B shows representative images for flow cytometry in N342 and S342macrophages, with the gating strategy of macrophages (inside panel) and the fluorescence intensity for Dio-LDL while panel C shows the results from five N342 and five S342 patients (mean intensity fluorescence isgiven, mean ± SEM is shown).

doi:10.1371/journal.pone.0122414.g003

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 9 / 12

expressed [7], we decided to investigate directly in primary human cells the impact of thers9370867 variant and therefore isolated PBMCs and generated macrophages from N342 andfrom S342 IDOL carriers.

IDOL mRNA expression was not different in N342 or S342 IDOL carriers and more impor-tantly LDL-R functionality, measured as the cell ability to uptake fluorescent labelled LDL wassimilar between N342 or S342 macrophages. These data obtained in primary cells further limit,also at the molecular levels, the impact of the rs9370867 variant on the IDOL gene. Of note, pa-tients with familial hypercholesterolaemia, when stratified according to the rs9370867 variant,had similar baseline plasma lipid levels but presented a different magnitude of LDL-C reduc-tion following statin therapy, suggesting that a pharmacogenetic effect for this IDOL SNPcould exist [29]. A similar finding was observed for another IDOL SNP (Rs6924995) in patientstreated with rosuvastatin [30]. In our cohort, NN342 carriers and SS342 carriers, when movedon statin therapy, showed a similar magnitude of LDL-C reduction (data not shown).

Based on these observations, the possibility that, in vitro, sterol deprivation followed byLXR stimulation could maximise the conditions to appreciate a difference, if any, on LDL up-take in carriers of the different isoforms should be considered.

However, it is unlikely, that these conditions occur in the general population and the clinicaldata point to a similar prevalence of cardiovascular disorders when stratifying for the rs9370867variant on the IDOL gene.

In conclusion, although IDOL represents a key player in LDL-R biology and subsequentrisk of dyslipidemia and atherosclerosis, the functional and clinical impact of the N342S variantis limited. The inclusion of this variant for improving genetic scores for hypercholesterolemiaor pharmacological approaches addressed at interfering with the region surrounding the aa342 of the enzyme should be carefully reconsidered.

Supporting InformationS1 Table. Clinical characteristics according to the rs9370867 SNP in subjects from thePLIC population included in the upper 75th percentile of total cholesterol (A) and triglycer-ides (B). -Multivariate Analysis.(DOCX)

AcknowledgmentsGDN is supported by Telethon Foundation (GGP13002) and Piano di Sviluppo UNIMI- LineaB 2013. ALC is supported by a grant from SISA Lombardia and SITeCS. None of this supportwas used directly for the study.

Author ContributionsConceived and designed the experiments: GDN ALC. Performed the experiments: AD GT ABKG. Analyzed the data: AD GT AB KG. Contributed reagents/materials/analysis tools: FP LG.Wrote the paper: AD GDN ALC.

References1. Catapano AL, Reiner Z, De Backer G, Graham I, Taskinen MR, Wiklund O, et al. ESC/EASGuidelines

for the management of dyslipidaemias The Task Force for the management of dyslipidaemias of theEuropean Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS). Atherosclero-sis 2011; 217: 3–46. PMID: 21882396

2. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science 1986;232: 34–47. PMID: 3513311

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 10 / 12

3. Hobbs HH, Russell DW, BrownMS, Goldstein JL. The LDL receptor locus in familial hypercholesterol-emia: mutational analysis of a membrane protein. Annu Rev Genet1990; 24: 133–170. PMID: 2088165

4. Goldstein JL, DeBose-Boyd RA, BrownMS. Protein sensors for membrane sterols. Cell 2006; 124:35–46. PMID: 16413480

5. Tibolla G, Norata GD, Artali R, Meneghetti F, Catapano AL. Proprotein convertase subtilisin/kexin type9 (PCSK9): from structure-function relation to therapeutic inhibition. Nutr Metab Cardiovasc Dis 2011;21: 835–843. doi: 10.1016/j.numecd.2011.06.002 PMID: 21943799

6. Norata GD, Tibolla G, Catapano AL. Targeting PCSK9 for hypercholesterolemia. Annu Rev PharmacolToxicol 2014; 54: 273–293. doi: 10.1146/annurev-pharmtox-011613-140025 PMID: 24160703

7. Zelcer N, Hong C, Boyadjian R, Tontonoz P. LXR regulates cholesterol uptake through Idol-dependentubiquitination of the LDL receptor. Science 2009; 325: 100–104. doi: 10.1126/science.1168974 PMID:19520913

8. Zhang L, Reue K, Fong LG, Young SG, Tontonoz P. Feedback regulation of cholesterol uptake by theLXR-IDOL-LDLR axis. Arterioscler Thromb Vasc Biol 2012; 32: 2541–2546. doi: 10.1161/ATVBAHA.112.250571 PMID: 22936343

9. Calkin AC, Lee SD, Kim J, van Stijn CM, Hong C, Tangirala RI, et al. Transgenic Expression of Domi-nant-Active IDOL in Liver Causes Diet-Induced Hypercholesterolemia and Atherosclerosis in Mice.Circ Res.2014; 115(4): 442–9. doi: 10.1161/CIRCRESAHA.115.304440 PMID: 24935961

10. Waterworth DM, Ricketts SL, Song K, Chen L, Zhao JH, Ripatti S, et al. Genetic variants influencing cir-culating lipid levels and risk of coronary artery disease. Arterioscler Thromb Vasc Biol 2010; 30: 2264–2276. doi: 10.1161/ATVBAHA.109.201020 PMID: 20864672

11. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, Koseki M, et al. Biological, clinicaland population relevance of 95 loci for blood lipids. Nature 2010; 466: 707–713. doi: 10.1038/nature09270 PMID: 20686565

12. Chasman DI, Pare G, Mora S, Hopewell JC, Peloso G, Clarke R, et al. Forty-three loci associated withplasma lipoprotein size, concentration, and cholesterol content in genome-wide analysis. PLoS Genet2009; 5: e1000730. doi: 10.1371/journal.pgen.1000730 PMID: 19936222

13. Weissglas-Volkov D, Calkin AC, Tusie-Luna T, Sinsheimer JS, Zelcer N, Riba L, et al. The N342SMYLIP polymorphism is associated with high total cholesterol and increased LDL receptor degradationin humans. J Clin Invest 2011; 121: 3062–3071. doi: 10.1172/JCI45504 PMID: 21765216

14. Santos PC, Oliveira TG, Lemos PA, Mill JG, Krieger JE, Pereira AC. MYLIP p.N342S polymorphism isnot associated with lipid profile in the Brazilian population. Lipids Health Dis2012; 11: 83. doi: 10.1186/1476-511X-11-83 PMID: 22741812

15. Sorrentino V, Fouchier SW, Motazacker MM, Nelson JK, Defesche JC, Dallinga-Thie GM. Identificationof a loss-of-function inducible degrader of the low-density lipoprotein receptor variant in individuals withlow circulating low-density lipoprotein. Eur Heart J 2013; 34: 1292–1297. doi: 10.1093/eurheartj/ehs472 PMID: 23324548

16. Lorenz MW, Polak JF, Kavousi M, Mathiesen EB, Volzke H, Tuomainen TP. Carotid intima-media thick-ness progression to predict cardiovascular events in the general population (the PROG-IMT collabora-tive project): a meta-analysis of individual participant data. Lancet 2012; 379: 2053–2062. doi: 10.1016/S0140-6736(12)60441-3 PMID: 22541275

17. Baragetti A, Knoflach M, Cuccovillo I, Grigore L, Casula M, Garlaschelli K, et al. Pentraxin 3 (PTX3)plasma levels and carotid intima media thickness progression in the general population. Nutr MetabCardiovasc Dis 2014; 24: 518–523. doi: 10.1016/j.numecd.2013.10.030 PMID: 24462365

18. Norata GD, Garlaschelli K, Grigore L, Raselli S, Tramontana S, Meneghetti F, et al. Effects of PCSK9variants on common carotid artery intima media thickness and relation to ApoE alleles. Atherosclerosis2010 208: 177–182. doi: 10.1016/j.atherosclerosis.2009.06.023 PMID: 19619878

19. Baragetti A, Norata GD, Sarcina C, Rastelli F, Grigore L, Garlaschelli K, et al. High density lipoproteincholesterol levels are an independent predictor of the progression of chronic kidney disease. J InternMed 2013; 274: 252–262. doi: 10.1111/joim.12081 PMID: 23607805

20. Norata GD, Ongari M, Garlaschelli K, Tibolla G, Grigore L, Raselli S, et al. Effect of the -420C/G variantof the resistin gene promoter on metabolic syndrome, obesity, myocardial infarction and kidney dys-function. J Intern Med 2007; 262: 104–112. PMID: 17598818

21. Ammirati E, Bozzolo EP, Contri R, Baragetti A, Palini AG, Cianflone D, et al. Cardiometabolic and im-mune factors associated with increased common carotid artery intima-media thickness and cardiovas-cular disease in patients with systemic lupus erythematosus. Nutr Metab Cardiovasc Dis 2014, 24:751–759. doi: 10.1016/j.numecd.2014.01.006 PMID: 24787906

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 11 / 12

22. Ammirati E, Cianflone D, Vecchio V, Banfi M, Vermi AC, De Metrio M, et al. Effector Memory T cells AreAssociatedWith Atherosclerosis in Humans and Animal Models. J Am Heart Assoc 2012, 1: 27–41.doi: 10.1161/JAHA.111.000125 PMID: 23130116

23. Baragetti A, Palmen J, Garlaschelli K, Grigore L, Pellegatta F, Tragi E, et al. Telomere shortening over6 years is associated with increased subclinical carotid vascular damage and worse cardiovascularprognosis in the general population. J Intern Med 2014, July 5. doi: 10.1111/joim.12282

24. Predazzi IM, Norata GD, Vecchione L, Garlaschelli K, Amati F, Grigore L, et al. Association betweenOLR1 K167N SNP and intima media thickness of the common carotid artery in the general population.PLoS One 2012; 7: e31086. doi: 10.1371/journal.pone.0031086 PMID: 22347434

25. Norata GD, Baragetti I, Raselli S, Stucchi A, Garlaschelli K, Vettoretti S, et al. Plasma adiponectin levelsin chronic kidney disease patients: relation with molecular inflammatory profile and metabolic status.Nutr Metab Cardiovasc Dis 2010; 20: 56–63. doi: 10.1016/j.numecd.2009.01.011 PMID: 19359150

26. Norata GD, Cattaneo P, Poletti A, Catapano AL. The androgen derivative 5alpha-androstane-3beta,17-beta-diol inhibits tumor necrosis factor alpha and lipopolysaccharide induced inflammatory response inhuman endothelial cells and in mice aorta. Atherosclerosis 2010; 212: 100–106. doi: 10.1016/j.atherosclerosis.2010.05.015 PMID: 20557886

27. Ferri N, Tibolla G, Pirillo A, Cipollone F, Mezzetti A, Pacia S, et al. Proprotein convertase subtilisinkexin type 9 (PCSK9) secreted by cultured smooth muscle cells reduces macrophages LDLR levels.Atherosclerosis 2012; 220: 381–386. doi: 10.1016/j.atherosclerosis.2011.11.026 PMID: 22176652

28. Yen CF, Kalunta CI, Chen FS, Kaptein JS, Lin CK, Lad PM, et al. Flow cytometric evaluation of LDL re-ceptors using DiI-LDL uptake and its application to B and T lymphocytic cell lines. J Immunol Methods1994; 177: 55–67. PMID: 7822838

29. Santos PC, Morgan AC, Jannes CE, Krieger JE, Santos RD, Pereira AC. The MYLIP p.N342S polymor-phism is associated with response to lipid-lowering therapy in Brazilian patients with familial hypercho-lesterolemia. Pharmacogenet Genomics 2014; 24: 548–555. doi: 10.1097/FPC.0000000000000089PMID: 25171759

30. Chasman DI, Giulianini F, MacFadyen J, Barratt BJ, Nyberg F, Ridker PM. Genetic determinants ofstatin-induced low-density lipoprotein cholesterol reduction: the Justification for the Use of Statins inPrevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial. Circ Cardiovasc Genet2012; 5: 257–264. doi: 10.1161/CIRCGENETICS.111.961144 PMID: 22331829

IDOL and Atherosclerosis

PLOS ONE | DOI:10.1371/journal.pone.0122414 April 30, 2015 12 / 12